| |

What 22 Years in the Amazon Revealed About a Quiet Climate Crisis

A long-term Amazon study shows how rising nighttime temperatures and wetter air are quietly reducing seed production, changing the forest from the inside out

I still remember the sound of the forest floor under my boots the first time I visited a tropical rainforest. The layers of fallen leaves, fruit husks, and seeds made the ground feel like a sponge. You don’t walk through that forest; you’re absorbed into it. 

Every few steps, there’d be a burst of movement: frogs leaping, leaves rustling from a tamarin, or seeds dropping from the canopy with that unmistakable plop that tells you a tree high above is doing its part to keep the forest going.

Working on conservation projects with local researchers and community leaders helps you understand rainforests at a whole new level. In recent years, though, many field stations have had to change their research projects and plan long-term biodiversity monitoring. This is partly to guide management but also because something is clearly shifting. 

The rains aren’t falling like they used to. Some tree species are flowering too early; others don’t fruit at all. People who had lived there all their lives noticed it. And so did we.

That’s why I was so struck by a recent study published in Ecology Letters by Dr. Jason Vleminckx and colleagues. Over 22 years, this team tracked seed production in Yasuní National Park in the Ecuadorian Amazon (see figure below), combining an impressive amount of plant reproductive data with climate records to see how the forest has responded to changing conditions. 

It’s one of the most detailed looks we have at what happens when climate change creeps into a place that, for so long, was thought to be buffered from it.

Location of the Yasuní Forest Dynamic Plot (FDP) within the bounds of Yasuní National Park (YNP; dark grey area) — Vleminckx et al. 2025

The researchers worked in Yasuní National Park in Ecuador, one of the wettest and most biodiverse spots in the world, and an area I was privileged to advise on during my tenure with conservation nonprofits. 

Rather than relying on assumptions or short-term trends, they spent more than two decades counting seeds from over 200 species using a network of seed traps. They matched those numbers with climate data: things like solar irradiance, temperature highs and lows, humidity, and rainfall. 

The goal? To figure out which climate variables most influence seed production, and whether those influences are direct (like heat damaging a developing fruit) or indirect (like flowers failing to appear in the first place).

The big picture is concerning. Seed production in this ever-wet forest has dropped over the past 22 years, not because of drought, fires, or deforestation. In fact, this forest is still largely intact. The drop is happening quietly, and it’s tied to two subtle but powerful trends: warmer nighttime temperatures and increasingly humid air.

Visual summary of the analysis linking climate variables to seed production. (a) Directed acyclic graph (DAG) showing how any climate variable © may influence seed production (SP) directly or indirectly through flower production (FP), with equations for total (T), direct (d), and mediated (m) effects. (b) Example using rainfall: solid and dashed black curves show mean monthly flower and seed production (scaled and centered); colored bars along the x-axis indicate months where rainfall had a significant positive (blue), negative (red), or non-significant (grey) direct effect on seed production. Thicker bars mark the strongest effects. © The 95% Bayesian credibility interval (shaded area) shows the linear relationship between seed production and rainfall’s direct effect, with solid and dashed lines indicating the significance and direction of direct and flower-mediated effects, respectively — Vleminckx et al. 2025

Here’s a more detailed explanation. At night, trees typically slow down and rest. But when nights get warmer, trees end up burning more of the sugars they made during the day just to keep their basic functions running. That’s energy that doesn’t go into making flowers or setting fruit. Over time, this adds up. 

The researchers found that rising minimum nighttime temperatures had a strong, negative effect on seed output, especially when those warm nights stretched through the flowering season.

On top of that, the forest air is getting wetter. That might sound like a good thing in the Amazon, but in this context (Western Amazon), it’s not. More humid air reduces the vapor pressure deficit (VPD), which is the driving force that lets water (and nutrients) move from roots to leaves. Lower VPD slows down that process, and again, reproduction takes a hit. 

The team found a clear pattern: as VPD went down, so did seed production. It turns out that changing conditions ain’t no good when you’re adapted to something different. 

Interestingly, though, not all climate effects were negative. Daytime heat had a mild positive effect, particularly when it came just before or during the fruiting season. It seems that warmer days can sometimes help. They hypothesized by speeding up metabolic processes or enhancing nutrient recycling as leaf litter breaks down. But that benefit wasn’t strong enough to counter the damage done by the warm, wet nights.

Monthly values (grey lines), interannual trends (black lines) and linear regression trends (dotted black lines) of (a) community-level seed production and each climate variable (b–f) between February 2000 (beginning of the flower and seed sampling) and August 2021 (end of the last fruiting phenological year used in the regression analyses). The interannual trends were obtained from Seasonal-Trend Decomposition using Loess (Cleveland et al. 1990). The blue-shaded areas in panel a indicate months where seed production data was imputed using an ARIMA model fitting — Vleminckx et al. 2025

Rainfall showed a more complex pattern. A bit more rain just before flowering seemed to help, possibly by providing a water buffer for the trees. But too much rain during fruiting? That’s a problem. 

Heavy rains can knock down developing fruits, spread fungal pathogens, or limit the light needed for certain processes. Timing really mattered here, and the authors’ careful statistical approach helped tease that apart.

But what does all this mean beyond the forest?

In practical terms, fewer seeds mean fewer chances for trees to regenerate. Over time, that can shift which species survive, how the forest looks, and which animals can live there. 

It’s easy to forget, but everything from birds to primates to insects depends on the pulses of fruit and seed that tropical trees produce. If that rhythm gets out of sync, the entire forest web stretches and sometimes tears.

Slope coefficients showing the total, direct, and flower-mediated effects of each climate variable on seed production, along with regression trends. In graphs (a–c, e–g, i–k, m–o, q–s), solid and dashed black lines represent average monthly seed and flower production (scaled from −1 to 1) over a 24-month phenological year. Blue, red, and grey bars indicate significantly positive, negative, or non-significant slope values, with thicker bars marking the strongest effects. Final column graphs (d, h, l, p, t) show the total (shaded), direct (solid), and mediated (dashed) effects with 95% credibility intervals; blue shading in panel (d) indicates a consistent positive trend despite non-significant slope coefficients — Vleminckx et al. 2025

It also challenges the idea that ever-wet tropical forests are somehow immune to climate change. For years, conservation plans have focused on drier forests, assuming the wetter ones would be more stable. But this study shows that even in one of the rainiest places on Earth, small shifts in climate, especially at night, can make a big difference.

As someone who’s spent years in the field, in labs, and around policy tables, I find this kind of work invaluable. Not just because of the science (which is rock solid), but because it helps connect dots between long-term monitoring, ecological function, and the climate changes happening around us.

We often talk about climate impacts in big, flashy terms such as fires, floods, and hurricanes. But sometimes, the biggest shifts start with something quiet. Like seeds that don’t fall. Or fruiting that comes a little later each year, until one day, it doesn’t come at all.

And it turns out, those subtle changes may carry some of the loudest warnings of all.


Do you want to read more stories like this? You can now subscribe to my newsletter and join a community of over 11,000 Earth lovers!

I’m thrilled to have you here. Stay curious, and thank you for being part of this journey!

Best,
Sílvia P-M, PhD Climate Ages

Join my mailing list!

Join my newsletter to explore how to transition from academia, build impactful side hustles, and communicate science effectively.

As a bonus, download my FREE Boosted Blog Method Cheat Sheet to kickstart your journey into science blogging and making an impact.

The boosted blog method teaser

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *